2026-09-22T18:02:40Zhttps://api.repository.cam.ac.uk/server/oai/request

oai:www.repository.cam.ac.uk:1810/3935072025-12-09T01:44:53Zcom_1810_224161com_1810_256067col_1810_224162

Epigenetic Heritability in Somatic Cells https://doi.org/10.17863/CAM.123824 Kregar, Lori Dolores Campbell, Peter Across the lifespan, cells accumulate somatic mutations, epigenetic alterations, tran- scriptional noise, mitochondrial damage and irreversible protein aggregation, the so-callled hallmarks of ageing. These molecular changes tend to worsen and compound over time, yet the mechanism through which these microscopic alterations in individual cells coalesce to entrain predictable, inevitable changes at the organ-wide level remains unclear. One way of explaining how stochastic damage at the cellular level could manifest in systemic changes of ageing is if clones with convergent phenotypes expand over time such that by old age these cells predominate. With age, normal tissues become colonised by expanding clones of cells that have acquired fitness-enhancing somatic mutations, so-called driver mutations, which enable them to over-proliferate, but it is unknown to what extent the other hallmarks of ageing can induce such behaviours. In this thesis I present my findings on epigenetic heritability in somatic cells. The thesis consists of four main chapters. In the first chapter, I describe Memento, a maximum likelihood method I developed for inferring ancestral methylation states. The method takes a phylogeny of somatic cells and methylation read counts of these cells to infer a zygotic starting state and which lineages incurred a methylation change to produce the observed methylation states in the cells. In the second chapter, I present the results of running Memento genome-wide on four healthy donors. I find that methylation at the majority of CpG sites is highly heritable and stably maintained for decades. I also establish there is an accelerated and transient period of epigenetic change in early development, occurring before gastrulation. Moreover, methylation changes are stochastic and tend to affect a single allele. Lastly, many of the methylation changes affect not just a single CpG site but entire clusters of adjacent CpG sites. In the third chapter, I study the role of methylation in cancer. I find that in breast cancer patients, hypermethylation of the BRCA1 promoter affects an ancestral lineage that existed in pre-gastrulation and gave rise to a sizeable proportion of the blood compartment as well as breast epithelial tissue in which, decades later, cancer develops. In the fourth chapter, I determine the timing of X inactivation and estimate the number of cells in an embryo at the moment each cell makes its choice of which parental copy to inactivate. In summary, I demonstrate that methylation is a stably inherited phenotype that is maintained through cell division, often for decades. Interestingly, I find that early embryogenesis is a crucial time for methylation profile establishment. During that critical time, there is rapid acquisition of enormous, genome-wide heterogeneity in DNA methylation - this seeds diverse cellular phenotypes that can decades later promote the evolution of cancer. In my thesis I show how causality of effects can be disentangled, which poses interesting questions about the role of methylation across diseases and ageing. University of Cambridge 2024-12-05 Thesis Doctoral Doctor of Philosophy (PhD) eng Wellcome Trust - 222382/Z/21/Z https://www.repository.cam.ac.uk/handle/1810/393507 embargo 2026-12-05 https://www.repository.cam.ac.uk/bitstreams/39d9b864-8519-4357-9ddf-1928b860daa1/download 08ad18f8552276ac056ae8de5d639d78 https://www.repository.cam.ac.uk/bitstreams/e67142bc-10f0-4e23-86c8-10067867fd12/download 87eda9de84448d1f82354d60eee3eb5f http://purl.org/NET/rdflicense/allrightsreserved epigenetics genomics methylation multiomics phylogenetics somatic mosaicism whole genome sequencing